Contextual Studies Methodology Research Paper

1. Historical Underpinnings

The contemporary study of context has its origins in a number of different disciplines, not only sociology and anthropology, but also ethology (the study of animal behavior in natural habitats), ecology (the study of biological organisms and their environment), and urban geography (the study of the development and functioning of neighborhoods in cities). All of these disciplines emphasize the fact that any human or animal behavior cannot be understood outside of the setting in which it occurs. In this sense, the organism and context are seen as inextricably linked.

Psychology, as a discipline, was slow to embrace a contextual point of view. Although numerous writers had attempted to move the field in this direction— most notably, Kurt Lewin (1951)—the study of psychology was dominated for much of the twentieth century by experimental scientists, who studied human behavior in the laboratory. Context was seen as a nuisance variable—something that got in the way of pure science—and the goal of experimental research was to study behavior under conditions that permitted the investigator to control any variable that was seen as extraneous to the research focus. If one wanted to study, for example, how individuals make decisions, the researcher would need to control the setting in which the decision-making took place, so that each subject in the study could be observed under identical circumstances. The simplest way to do this was to perform research in a laboratory, where each and every element of the context could be controlled.

The benefit of controlled laboratory research is that it permits the scientist to isolate pure behavioral phenomena, uncontaminated by the outside environment. But this benefit must be weighed against the fact that phenomena observed under controlled experimental conditions may not be the same as phenomena observed in their natural context. The way in which individuals make decisions in a psychology laboratory, for instance, may not resemble the way they make decisions at home, at work, or in a store. Just as ethologists would not want to generalize observations of caged animals to the behavior of that same species in the wild, social scientists must be careful about generalizing findings derived from controlled experiments to human behavior outside the laboratory. This is not to question the importance of laboratory-based research but, rather, to note that the laboratory is itself a context, and that features of this context, like any other context, necessarily influence the behavior that occurs within it.

During the 1970s, concerns about ecological validity—the extent to which an observed phenomenon in a research investigation is a valid reflection of that phenomenon as it occurs in natural contexts—were voiced by numerous investigators working in a variety of areas. In the study of intelligence, for example, scholars raised questions about whether standardized intelligence tests adequately captured intelligent behavior as it occurs in the real world. Several researchers found that individuals who performed poorly on standardized tests of mathematics frequently demonstrated the ability to use very sophisticated mathematical reasoning when observed during day-to-day situations, such as shopping. Similarly, scholars of social behavior discovered that individuals’ behavior was considerably influenced by the context in which it took place, and that conclusions drawn from studies of behavior in one context (e.g., observations of young children at home) could not always be extended to other settings (e.g., observations of these same youngsters in the preschool classroom). Ecological psychologists drew attention to the fact that individual behavior was profoundly shaped by the structure of the physical environment in which people lived. Several classic studies of schools indicated, for ex-ample, that students’ behavior in small schools was markedly different from students’ behavior in large schools (Barker and Gump 1964).

In the late 1970s, psychologist Urie Bronfenbrenner drew widespread attention to the role of context in human development and behavior by noting that the scientific literature on child development was based almost entirely on the study of children interacting with strangers in strange situations for the briefest amounts of time. How, Bronfenbrenner wondered, could social scientists hope to contribute to the solution of social problems if they insisted on studying human behavior in a contextual vacuum? Although his comments were aimed at students of child development, they applied equally well to other disciplines. In his book, The Ecology of Human Development, Bronfenbrenner (1979) called on the field to think in more sophisticated fashion about the settings in which development took place, and he outlined a model of context that researchers could use to guide their thinking and empirical investigation. This work, and the thinking it would subsequently inspire, led to the development of the ecological perspective on human development, a perspective that grew more and more influential over the next several decades (Bronfenbrenner and Crouter 1983).

2. The Ecological Perspective

Within the ecological perspective, the context is construed as a series of nested environments, with each level of contextual influence embedded within a larger level, not unlike a set of Russian dolls. At the innermost, or most proximal level of contextual influence, are the immediate settings in which individuals spend time—the family, the friendship group, the classroom, the workplace, the laboratory, and so forth. Most studies of the role of context in shaping human behavior and development are studies of single contexts examined in isolation from one another (e.g., the impact of the family context on child development, the impact of teaching practices on student achievement).

The ecological perspective provides a conceptual framework within which to investigate more complex interactions between persons and environments, how-ever. Three notions are especially important in this regard. First, proximal settings are seen as connected to each other, because events that take place in one setting often have ramifications for individual behavior and development in another. Socialization in the family context influences how adolescents behave in the peer group, peer influence processes affect how adolescents behave in the classroom, experiences in the classroom affect students’ performance in extra-curricular settings, and so on. Within the ecological perspective, therefore, an important focus of inquiry is on the ways in which, and the extent to which, different settings are linked, and the impact of these links on individual behavior.

Second, within the ecological perspective, these proximal settings are seen as contained within broader institutional and community contexts that shape the structure of settings and influence what takes place within them (Duncan and Raudenbush 1998, Jencks and Mayer 1990). Families and friendship networks reside within neighborhoods, for example, and qualities of neighborhoods influence how families and friendship networks function. Parents interact with their children differently in poor neighborhoods than they do in more affluent ones, for example, and, as a consequence, child development may vary across neighborhood contexts. Similarly, schools exist within communities, and characteristics of communities— urban vs. rural, tightly knit vs. impersonal—necessarily influence the ways in which schools are organized and operate. Within the ecological perspective, therefore, one needs to examine relations among nested contexts as well as between individuals and settings.

Finally, proximal settings and the broader environments that contain them are located within particular historical, social, economic, political, geographical, and cultural contexts—‘macro-systems’—that influence the nature, structure, function, organization, and influence of all levels of the environment. Schools in Japan function differently than those in Canada, and differences between Japanese and Canadian schools have implications for the ways that teachers and students interact with each other, which in turn have implications for understanding student achievement in each context. A city’s neighborhoods today may bear little resemblance to these very same neighbor-hoods 50 years ago, and the ways in which neighbor-hoods have changed over historical time may have important implications for understanding transformations in family life and parent-child relationships. Important social and economic events—wars, economic depressions, natural disasters—shape proximal and distal settings in important ways that have ramifications for individuals’ behavior and development. Within the ecological perspective, researchers are interested in the influence of macro-systems on the settings located within them, and, consequently, in the rippling effects of these influences on individual behavior and development.

3. Methodological Challenges In Contextual Research

As more and more social scientists began to incorporate an increasingly sophisticated model of context into their research designs, it became clear that the scientific study of context presented a number of new and formidable methodological challenges. The discussion shifted from whether one needed to study context (the consensus was that one did) to how best to do so. Among the most important of these concerns involved deciding upon the most appropriate way to assess context, developing models of the transactional nature of contextual influence, and disentangling the effects of hierarchically organized, or nested, environments.

3.1 Objective vs. Subject Assessment Of Context

One of the longstanding debates in the study of context concerned the relative merits of objective vs. subjective assessments of the environment. Objective assessments attempt to separate the measurement of context from the assessment of the individuals whose behavior is the target of investigation. A researcher interested in the influence of the parent–child relationship on child development might conduct impartial observations of family interaction in the home and draw his or her own conclusions about the relationship on the basis of what was observed. Similarly, a student of neighbor-hood stress and its impact on mental health might draw upon official statistics on crime, poverty, un-employment, and housing density and look at these in relation to residents’ mental health problems.

Subjective assessments of context, in contrast, rely on the research participants’ perceptions of their environment. These perceptions can be assessed by asking individuals to complete questionnaires or by interviewing them about the settings in which they live. Thus, a researcher interested in parent–child relationships might interview parents about their socialization practices or ask children to complete questionnaires about their parents’ behavior. Similarly, someone interested in neighborhood stress and mental health might collect information from neighborhood residents on their perceptions of the com-munity’s dangerousness, economic health, or housing quality and examine the links between these perceptions and one or more indices of mental health.

These different approaches to the assessment of context have particular strengths and weaknesses. The advantage of using objective assessments inheres in the fact that the assessment of context is not contaminated by the characteristics of the individuals whose behavior is being studied. Suppose, for instance, a researcher interested in neighborhood stress and mental health gathered information about each by interviewing a sample of residents and found that depression was more prevalent in dangerous areas. Because depressed individuals may describe their neighborhoods in more negative terms than nondepressed individuals, however, it would be difficult to know whether the observed link between dangerous-ness and depression was genuine or merely an artifact of biased reporting. If, however, the relationship between dangerousness and depression were con-firmed in a study using official crime statistics, one would have more confidence in the validity of the finding.

The advantage of using subjective assessments of context is that such assessments capture the context as it is experienced by the participant. Some social scientists argue that it is the subjectively experienced context, and not the objective context, that has the most profound impact on behavior and development. Individuals’ perceptions of their neighborhood as safe vs. dangerous, for example, may well influence how they behave far more than the actual level of crime as measured by official statistics. As the sociologist W. I. Thomas wrote early in the twentieth century, if individuals ‘perceive situations as real, they are real in their consequences’ (Thomas and Thomas 1928). Of course, one problem that arises when researchers rely on subjective assessments of context concerns the reconciliation of different subjects’ perceptions of the same context. In family research, for example, it is common to find that different family members characterize their family context in different ways.

Ultimately, the most thorough assessment of con-text is one that employs multiple sources of information and uses multiple methods of assessment. By combining information from objective and subjective sources, from multiple vantage points, and gained through observational, self-report, and archival data, researchers are often able to provide richer and more accurate characterizations of a setting than are achieved through one approach alone.

3.2 The Transactional Nature Of Contextual Influence

A second methodological challenge involves capturing the transactional nature of contextual influence. Although researchers often claim to be studying the influence of context on the individual, in reality the situation is far more complicated than this, in at least three important ways. First, individuals are not the passive recipients of contextual influence, but are active agents who shape the contexts in which they live. At the same time, contexts are not static entities, but settings that change and develop over time. Thus, while children are no doubt influenced by their family environment, family environments are themselves influenced by the children who inhabit them. Similarly, friendship groups both affect, and are affected by, their members. Finding ways to model the reciprocal and dynamic relations between individuals and their environments has been an especially difficult challenge for researchers.

Second, it is often difficult to separate the effects of contextual selection from those of contextual influence (Jencks and Mayer 1990). In laboratory experiments individuals can be randomly assigned by the researcher to different groups and then compared to see whether group membership affects individual behavior. Out-side the laboratory, however, people often pick the contexts in which they spend time. Consider, for example, the often reported finding that delinquent teenagers often are members of peer groups composed of other delinquent adolescents. Is this because delinquent peers influence their friends to engage in antisocial behavior, or is it due, instead, to the propensity of delinquent individuals to choose to affiliate with others who share similar attitudes and values? Obviously, it is not possible to answer this question without randomly assigning individuals to different peer groups (not a feasible option in most research) or observing individuals and contexts over time. Similar problems exist in attempts to document the influence of neighborhoods on their residents. Because individuals have some control over where they live, it is difficult to disentangle the effects that neighborhoods have on individuals from the factors that lead some people to select particular neighbor-hoods in which to settle.

Third, different individuals are often affected by the same contextual factors in very different ways, making it extremely difficult to generalize about the effects of various contextual factors on individual behavior and development. A highly active and impulsive child may fare poorly in a relatively unstructured classroom, whereas this same context may be perfectly suitable for a more passive, more placid youngster. Similarly, an employee who is self-reliant and self-directed may thrive in a work environment that permits a great deal of autonomy, but this same work environment may prove problematic for an employee who needs a great deal of guidance and supervision. The recognition that contextual influence varies as a function of differences between individuals has stimulated a good deal of interest in the study of ‘person-environment fit’—the extent to which a given context meshes well with the individual’s personal characteristics.

3.3 Hierarchical Nature Of Contextual Influence

The third methodological challenge concerns the hierarchical and interactive nature of contextual influence. The recognition that individuals are affected by multiple contexts simultaneously has necessitated the development of more sophisticated conceptual and statistical models in order to best describe the complicated ways through which individuals are influenced by the contexts in which they live (Duncan and Raudenbush 1998). Some contexts, for example, affect individuals directly, as when the behavior of a parent directly affects the development of a child. Other contextual effects occur through their impact on some other setting, however; stress in the parent’s workplace may affect the child indirectly, through its impact on the family. In this instance, the second setting (in this case, the family) is said to ‘mediate’ the impact of the first setting (i.e., the workplace). And still other contextual effects occur by transforming, or ‘moderating,’ the impact of other contexts. Neighbor-hoods, for example, often exert their strongest impact on individuals not directly, but by attenuating or enhancing the impact of other contexts, such as the family or school. Within the context of an affluent and safe neighborhood, close parental monitoring may have deleterious effects on the child (because such monitoring may be experienced by the child as overprotective or intrusive), but the very same level of parental monitoring may have beneficial effects within the context of a dangerous, economically stressed community. In order to analyze direct, mediating, and moderating effects, researchers have had to develop new classes of multivariate statistical procedures, such as path analysis, structural equation modeling, and hierarchical linear modeling.

4. Concluding Note

The growing appreciation of the importance of context in the study of human behavior and development stimulated a wave of new research in the 1980s and 1990s that sought to understand how individuals are affected by different levels of the environment, both independently and in interaction with each other. Multidisciplinary and interdisciplinary research that brought together social scientists from different disciplines became increasingly common, as scholars realized that understanding the complex interrelations among individuals and their environments necessitated drawing on the methods and theories of multiple disciplines. Over time, the study of individuals in natural settings has come to dominate research in fields in which controlled laboratory experiments had once been the dominant paradigm. Once considered a nuisance that interfered with the pursuit of ‘pure knowledge,’ the context of human behavior is now recognized as a necessary and enormously important focus of contemporary social science inquiry.

Bibliography:

1. Barker R G, Gump P V 1964 Big School, Small School: High School Size and Student Behavior. Stanford University Press, Stanford, CA
2. Bronfenbrenner U 1979 The Ecology of Human Development: Experiments By Nature and Design. Harvard University Press, Cambridge, MA
3. Bronfenbrenner U, Crouter A 1983 The evolution of environ-mental models in developmental research. In: Kessen W, Mussen P H (eds.) Handbook of Child Psychology; History, Theory, and Methods, 4th edn., Wiley, New York, Vol. 1, pp. 357–414
4. Duncan G, Raudenbush S 1998 Neighborhoods and Adolescent Development: How can we Determine the Links? Working Paper Series, Institute for Policy Research, Northwestern University, Evanston, IL
5. Jencks C, Mayer S 1990 The social consequences of growing up in a poor neighborhood. In: Lynn L E, McGeary M G (eds.) Inner-City Poverty in the United States, National Academy Press, Washington, DC pp. 111–86
6. Lewin K 1951 Field Theory and Social Science. Harper, New York
7. Thomas W I, Thomas D S 1928 The Child in America. Knopf, New York